<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Finally: STB SPINVOX REVIEW</title>
	<atom:link href="http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/feed" rel="self" type="application/rss+xml" />
	<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review</link>
	<description></description>
	<lastBuildDate>Wed, 28 Jul 2010 16:18:54 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Igor Jablokov</title>
		<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/comment-page-1#comment-1567</link>
		<dc:creator>Igor Jablokov</dc:creator>
		<pubDate>Thu, 04 Jun 2009 00:48:17 +0000</pubDate>
		<guid isPermaLink="false">http://www.speechtechblog.com/?p=1781#comment-1567</guid>
		<description>Don&#039;t worry about reviewing us directly, as you&#039;ll probably end up being exposed to us elsewhere as we already power a number of the existing voicemail-to-text providers with our automation. Since our founding, we&#039;ve evolved to become a platform provider rather than a direct to consumer offering (we can provide functionality similar to Google Search or Google Voice as a white label solution). Some of the same R&amp;D staff that developed Dragon and ViaVoice are working their magic herein. Cheers!</description>
		<content:encoded><![CDATA[<p>Don&#8217;t worry about reviewing us directly, as you&#8217;ll probably end up being exposed to us elsewhere as we already power a number of the existing voicemail-to-text providers with our automation. Since our founding, we&#8217;ve evolved to become a platform provider rather than a direct to consumer offering (we can provide functionality similar to Google Search or Google Voice as a white label solution). Some of the same R&amp;D staff that developed Dragon and ViaVoice are working their magic herein. Cheers!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eric B.</title>
		<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/comment-page-1#comment-1510</link>
		<dc:creator>Eric B.</dc:creator>
		<pubDate>Tue, 02 Jun 2009 16:47:06 +0000</pubDate>
		<guid isPermaLink="false">http://www.speechtechblog.com/?p=1781#comment-1510</guid>
		<description>I didn&#039;t have a post on Phone Tag. They just contacted me to see if I was interested in reviewing them.

http://www.speechtechblog.com/2009/04/14/spinvox-review-a-coming-to-stb#comments

I&#039;ll probably get around to it sooner or later.</description>
		<content:encoded><![CDATA[<p>I didn&#8217;t have a post on Phone Tag. They just contacted me to see if I was interested in reviewing them.</p>
<p><a href="http://www.speechtechblog.com/2009/04/14/spinvox-review-a-coming-to-stb#comments" rel="nofollow">http://www.speechtechblog.com/2009/04/14/spinvox-review-a-coming-to-stb#comments</a></p>
<p>I&#8217;ll probably get around to it sooner or later.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Inci</title>
		<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/comment-page-1#comment-1502</link>
		<dc:creator>Inci</dc:creator>
		<pubDate>Mon, 01 Jun 2009 21:09:24 +0000</pubDate>
		<guid isPermaLink="false">http://www.speechtechblog.com/?p=1781#comment-1502</guid>
		<description>Phonetag reached at to me as you can see in the comments of my earlier post. 

Eric, could you please guide me as to where your eariler post on Phonetag is?

Thanks,
Inci</description>
		<content:encoded><![CDATA[<p>Phonetag reached at to me as you can see in the comments of my earlier post. </p>
<p>Eric, could you please guide me as to where your eariler post on Phonetag is?</p>
<p>Thanks,<br />
Inci</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eric B.</title>
		<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/comment-page-1#comment-1499</link>
		<dc:creator>Eric B.</dc:creator>
		<pubDate>Sat, 30 May 2009 13:00:09 +0000</pubDate>
		<guid isPermaLink="false">http://www.speechtechblog.com/?p=1781#comment-1499</guid>
		<description>Inci: Phonetag reached at to me as you can see in the comments of my earlier post. However, as of yet I&#039;ve made no plans to review any further services. It&#039;s a long and tedious process, and I think I&#039;m looking forward to a break of just getting my voicemails the old fashioned way. I may take one of them up later, though.

Philippe: I understand your concern about SpinVox not actually having ASR. I&#039;ve seen that before. I think a lot of that sentiment comes from a patent that the company filed in 2006 that seems to lay out a transcription service manned by breathing human transcriptionists. It&#039;s still pending. You can find it here: 

http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&amp;Sect2=HITOFF&amp;d=PG01&amp;p=1&amp;u=%2Fnetahtml%2FPTO%2Fsrchnum.html&amp;r=1&amp;f=G&amp;l=50&amp;s1=%2220060223502%22.PGNR.&amp;OS=DN/20060223502&amp;RS=DN/20060223502

However, a series of patents later filed from 2007-2008 (which seem to be revisions of each other--though I&#039;m not an attorney) suggest that SpinVox is using ASR first and then having human transcriptionists make corrections.

Here&#039;s the most recent filing:

http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&amp;Sect2=HITOFF&amp;p=1&amp;u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&amp;r=1&amp;f=G&amp;l=50&amp;co1=AND&amp;d=PG01&amp;s1=Spinvox.AS.&amp;OS=AN/Spinvox&amp;RS=AN/Spinvox

Likewise, SpinVox has developed an API:

http://www.speechtechmag.com/Articles/News/Industry-News/SpinVox-Launches-Open-Access-to-Its-Speech-Platform-with-Web-API-53274.aspx

Seems strange that they would do that without some underlying recognition technology.

Plus, they entered into a partnership with IBM in April to do STT as a managed service on IBM hardware:

http://www.speechtechmag.com/Articles/News/Industry-News/SpinVox-Launches-Open-Access-to-Its-Speech-Platform-with-Web-API-53274.aspx

A foolhardy move on IBM&#039;s part if SpinVox is just a room full of minimum wage slaves sweatin&#039; out the transcriptions in some darkened room in some distant land, as some have seemed to suggest.

Apart from the company&#039;s own explanation, ( http://www.spinvox.com/how_it_works.html ) I just think there&#039;s too much outside, verifiable evidence to suggest they ARE using ASR at some level than there is to suggest are not using it at all.

All that said, from what I&#039;ve seen in the trial run, the engine is very much voicemail-to-text-centered. I don&#039;t know if it&#039;s strong enough to have application outside of that. Certainly from a dictation standpoint, Dragon seemed to out preform it. SpinVox, for its part, may be developing, or maybe even has developed, a different engine for purposes outside of voicemail. Who the heck knows?

I appreciate the skepticism, though.</description>
		<content:encoded><![CDATA[<p>Inci: Phonetag reached at to me as you can see in the comments of my earlier post. However, as of yet I&#8217;ve made no plans to review any further services. It&#8217;s a long and tedious process, and I think I&#8217;m looking forward to a break of just getting my voicemails the old fashioned way. I may take one of them up later, though.</p>
<p>Philippe: I understand your concern about SpinVox not actually having ASR. I&#8217;ve seen that before. I think a lot of that sentiment comes from a patent that the company filed in 2006 that seems to lay out a transcription service manned by breathing human transcriptionists. It&#8217;s still pending. You can find it here: </p>
<p><a href="http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&#038;Sect2=HITOFF&#038;d=PG01&#038;p=1&#038;u=%2Fnetahtml%2FPTO%2Fsrchnum.html&#038;r=1&#038;f=G&#038;l=50&#038;s1=%2220060223502%22.PGNR.&#038;OS=DN/20060223502&#038;RS=DN/20060223502" rel="nofollow">http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&#038;Sect2=HITOFF&#038;d=PG01&#038;p=1&#038;u=%2Fnetahtml%2FPTO%2Fsrchnum.html&#038;r=1&#038;f=G&#038;l=50&#038;s1=%2220060223502%22.PGNR.&#038;OS=DN/20060223502&#038;RS=DN/20060223502</a></p>
<p>However, a series of patents later filed from 2007-2008 (which seem to be revisions of each other&#8211;though I&#8217;m not an attorney) suggest that SpinVox is using ASR first and then having human transcriptionists make corrections.</p>
<p>Here&#8217;s the most recent filing:</p>
<p><a href="http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&#038;Sect2=HITOFF&#038;p=1&#038;u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&#038;r=1&#038;f=G&#038;l=50&#038;co1=AND&#038;d=PG01&#038;s1=Spinvox.AS.&#038;OS=AN/Spinvox&#038;RS=AN/Spinvox" rel="nofollow">http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&#038;Sect2=HITOFF&#038;p=1&#038;u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&#038;r=1&#038;f=G&#038;l=50&#038;co1=AND&#038;d=PG01&#038;s1=Spinvox.AS.&#038;OS=AN/Spinvox&#038;RS=AN/Spinvox</a></p>
<p>Likewise, SpinVox has developed an API:</p>
<p><a href="http://www.speechtechmag.com/Articles/News/Industry-News/SpinVox-Launches-Open-Access-to-Its-Speech-Platform-with-Web-API-53274.aspx" rel="nofollow">http://www.speechtechmag.com/Articles/News/Industry-News/SpinVox-Launches-Open-Access-to-Its-Speech-Platform-with-Web-API-53274.aspx</a></p>
<p>Seems strange that they would do that without some underlying recognition technology.</p>
<p>Plus, they entered into a partnership with IBM in April to do STT as a managed service on IBM hardware:</p>
<p><a href="http://www.speechtechmag.com/Articles/News/Industry-News/SpinVox-Launches-Open-Access-to-Its-Speech-Platform-with-Web-API-53274.aspx" rel="nofollow">http://www.speechtechmag.com/Articles/News/Industry-News/SpinVox-Launches-Open-Access-to-Its-Speech-Platform-with-Web-API-53274.aspx</a></p>
<p>A foolhardy move on IBM&#8217;s part if SpinVox is just a room full of minimum wage slaves sweatin&#8217; out the transcriptions in some darkened room in some distant land, as some have seemed to suggest.</p>
<p>Apart from the company&#8217;s own explanation, ( <a href="http://www.spinvox.com/how_it_works.html" rel="nofollow">http://www.spinvox.com/how_it_works.html</a> ) I just think there&#8217;s too much outside, verifiable evidence to suggest they ARE using ASR at some level than there is to suggest are not using it at all.</p>
<p>All that said, from what I&#8217;ve seen in the trial run, the engine is very much voicemail-to-text-centered. I don&#8217;t know if it&#8217;s strong enough to have application outside of that. Certainly from a dictation standpoint, Dragon seemed to out preform it. SpinVox, for its part, may be developing, or maybe even has developed, a different engine for purposes outside of voicemail. Who the heck knows?</p>
<p>I appreciate the skepticism, though.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: philippe</title>
		<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/comment-page-1#comment-1498</link>
		<dc:creator>philippe</dc:creator>
		<pubDate>Sat, 30 May 2009 06:43:24 +0000</pubDate>
		<guid isPermaLink="false">http://www.speechtechblog.com/?p=1781#comment-1498</guid>
		<description>Are you REALLY sure that spinvox is ASR based ?
I wonder why no one put a &quot;powered by&quot; to gain more visibility, having developped such a great engine ...</description>
		<content:encoded><![CDATA[<p>Are you REALLY sure that spinvox is ASR based ?<br />
I wonder why no one put a &#8220;powered by&#8221; to gain more visibility, having developped such a great engine &#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Inci</title>
		<link>http://www.speechtechblog.com/2009/05/29/finally-stb-spinvox-review/comment-page-1#comment-1495</link>
		<dc:creator>Inci</dc:creator>
		<pubDate>Sat, 30 May 2009 02:13:03 +0000</pubDate>
		<guid isPermaLink="false">http://www.speechtechblog.com/?p=1781#comment-1495</guid>
		<description>Thanks for the Nuance VM2T v SpinVox reviews. Will you be reviewing Yap, Vlingo and Promptu&#039;s Shoutout for the iPhone as well (i.e. when the latter is available)?  My understanding is that these three do not have human oversight thereby reflecting a more accurate performance of the ASR built for SMS, email, and voicemail, (i.e. speech to text (STT)). How about Phonetag? Don&#039;t know if the latter has human agents involved.</description>
		<content:encoded><![CDATA[<p>Thanks for the Nuance VM2T v SpinVox reviews. Will you be reviewing Yap, Vlingo and Promptu&#8217;s Shoutout for the iPhone as well (i.e. when the latter is available)?  My understanding is that these three do not have human oversight thereby reflecting a more accurate performance of the ASR built for SMS, email, and voicemail, (i.e. speech to text (STT)). How about Phonetag? Don&#8217;t know if the latter has human agents involved.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
